Gene Mutational Analyses In 154 Ovarian Cancer (Oc) Samples From The Rosia Study Of Front-Line Bevacizumab (Bev)-Containing Therapy For Oc

Background: In the single-arm ROSiA study (NCT01239732), 1021 patients (pts) with newly diagnosed OC received 12–24 wks of carboplatin + paclitaxel with BEV for up to 24 mo or until progression [Oza 2016]. Progression-free survival (PFS) was a secondary endpoint. Methods: In an optional translational research substudy, tumour tissue samples collected before BEV were analysed using Foundation Medicine Inc.’s FoundationOne® (FMOne) gene panel. Prevalence of gene alterations, tumour mutational burden (TMB) and potential prognostic effects were assessed in exploratory gene mutational analyses using Cox proportional hazards models. Correlations between TMB and BRCA1 mutation and immunohistochemical expression of the immune markers PD-L1 and CD8 were assessed. Results: The FMOne population (n = 154) was representative of the ITT population (n = 1021) for baseline characteristics but had slightly more favourable PFS (median 30.2 vs 25.5 mo, respectively; hazard ratio [HR] 1.16 [95% CI 0.92–1.45]; p = 0.21). The most common gene alterations (predominantly short variant) were TP53 (79% of pts), BRCA1 (22%), NF1 (14%), KRAS (10%), PIK3CA (9%) and BRCA2 (7%). MYC, TERC, GNAS and CCNE1 were amplified in 25%, 16%, 7% and 7%, respectively. The mean TMB (excluding germline polymorphisms and known cancer drivers) was 4.5/megabase (Mb; range 0–43). Only 2 pts had a TMB >15/Mb; 43 had a TMB ≥5/Mb. In univariate Cox regression analyses, none of the mutations explored showed a clear association with PFS. PFS slightly (not statistically significantly) favoured pts with (n = 34) vs without (n = 118) BRCA1 mutation (median 30.4 vs 28.1 mo, respectively; HR 0.76 [95% CI 0.44–1.31]; p = 0.32). TMB and PFS showed no association. No meaningful correlations were seen between either TMB or BRCA1 mutation and expression of the immune markers PD-L1 and CD8 on immune cells. Conclusions: Samples from pts with newly diagnosed OC indicated relatively infrequent gene alterations and low TMB. None of the gene alterations evaluated suggested prognostic value, but low frequency of these mutations and the relatively small number of samples in ROSiA limit interpretation, particularly for the correlation analyses. Clinical trial identification: NCT01239732 Legal entity responsible for the study: F Hoffmann-La Roche Ltd Funding: F Hoffmann-La Roche Ltd Disclosure: R. Deurloo, Y. Ghazi: Employee, shares: F Hoffmann-La Roche. N. Colombo: Travel expenses from Roche for ROSiA steering committee meetings, personal fees from AstraZeneca and PharmaMar for advisory boards and speaker engagements, and personal fees from Clovis, Amgen, MSD, and Pfizer for advisory boards. C. Mendiola: Non-financial support from Roche Farma Spain and AstraZeneca Spain. F. Selle: Honoraria for consultant/advisory role: Roche; travel/accommodation expenses: Roche, AstraZeneca. M.Z. Oestergaard, H-J. Urban: Employee: F Hoffmann-La Roche. M. Bylesjo: Employee of Fios Genomics Ltd, a contract research organisation contracted to provide statistical services to F Hoffmann-La Roche Ltd. A. Oza: Travel expenses from Roche for advisory boards and ROSiA steering committee meetings. All other authors have declared no conflicts of interest.